Homogeneous Conversion of Methane to Methanol. 2. Catalytic

J. Am. Chem. Soc. 2000, 122, 2041-2052

2041

Homogeneous Conversion of Methane to Methanol. 2. Catalytic Activation of Methane by cis- and trans-Platin: A Density Functional Study of the Shilov Type Reaction Kausala Mylvaganam,† George B. Bacskay,† and Noel S. Hush*,†,‡ Contribution from the School of Chemistry, UniVersity of Sydney, NSW 2006, Australia ReceiVed August 19, 1999. ReVised Manuscript ReceiVed NoVember 29, 1999

Abstract: The C-H activation of methane catalyzed by cis- and trans-platin in aqueous solution has been studied by density functional based computational methods. By analogy with the Shilov reaction, the initial step is the replacement of an ammonia ligand by methane, followed by the formation of a methyl complex and the elimination of a proton. The computations utilize the B3LYP hybrid functionals, effective core potentials, and double-ζ to polarized double-ζ basis sets and include solvation effects by a dielectric continuum method. In contrast with the Shilov reaction studied by Siegbahn and Crabtree (J. Am. Chem. Soc. 1996, 118, 4442), in the platins the replacement of an ammonia ligand by methane is found to be effectively rate determining, in that the energy barriers to C-H activation are comparable with those of the initial substitution reaction, viz. ∼ 34 and 44 kcal/mol for cis- and trans-platin, respectively. Several reaction pathways for C-H activation and subsequent proton elimination were identified. For cis-platin the energy barriers associated with the oxidative addition and σ-bond metathesis type mechanisms were found to be comparable, while for trans-platin oxidative addition is predicted to be strongly preferred over σ-bond metathesis, which, interestingly, also proceeds through a Pt(IV) methyl hydrido complex as reaction intermediate. In line with accepted ideas on trans influence, the methyl and hydride ligands in the Pt(IV) complexes that arise in the oxidative addition reactions were always found to be cis to each other. On the basis of the population analyses on the Pt(IV) complexes it is suggested that the Pt-H and Pt-CH3 bonds are best described as covalent bonds and, further, that the preference of the hydride and methyl anions to be cis to each other is a consequence of such covalent bonding. In light of these findings, the energies of several methyl Pt(IV) hydrido bisulfate complexes were also recalculated, with CH3 and H placed cis to each other. The revised results provide evidence for the thermodynamic feasibility of oxidative addition of methane to catalysts such as [Pt(NH3)2(OSO3H)2] or [Pt(NH3)2(OSO3H)(H2SO4)]+.

Introduction The catalytic conversion of methane to methanol is a problem of considerable scientific and economic importance. Methane, currently our cheapest and most abundant natural hydrocarbon resource, is underutilized by industry because of its high transport and storage costs. Most current technologies convert methane to methanol, which is already a major feedstock in the chemical industry, via the generation of “syngas” (CO + H2), whereby methane is fully oxidized at a high temperature. The potential benefits of an economical low-temperature alternative to the syngas technology are obviously great, and as a consequence, much current research is devoted to this problem. As the direct oxidation of methane to methanol involves the activation and functionalization of only one of the C-H bonds of methane, the obvious challenge is to find catalysts that could enable these reactions to proceed at reasonably low temperatures. The oxidation of methane by platinum salts in solution, i.e., via homogeneous catalysis, was first reported in the 1970s by Shilov and co-workers, and by now there is a considerable body of literature on the subject that report a range of transition metal †

School of Chemistry, University of Sydney. Department of Biochemistry, University of Sydney, NSW 2006, Australia (1) Arndtsen, B. A.; Bergman, R. G.; Mobley, T. A.; Peterson, T. H. Acc. Chem. Res. 1995, 28, 154. ‡

compounds that catalyze oxidation.1-10 The mechanism proposed by Shilov et al.5,7 for alkane oxidation consists of three basic steps: (a) activation of the alkane by a Pt(II) complex producing an alkyl Pt(II) complex which (b) is oxidized to a Pt(IV) alkyl intermediate that (c) will undergo reductive elimination, yielding the oxidized alkane (alkyl alcohol, ester, halide, etc.) and the original Pt(II) catalyst. Unfortunately, thus far, commercial developments have been hampered by low yields, slow kinetics, and the poisoning of the catalysts. Recently, Periana et al.11 reported the successful activation and functionalization of methane in H2SO4, by bis(2,2′-bipyrimidene)PtIICl2, that resulted in a 70% one pass yield of methanol. This followed their earlier work that explored the (2) Davies, J. A., Watson, P. L.; Liebman, J. L., Greenberg, A., Eds. SelectiVe Hydrocarbon ActiVation; Wiley-VCH: New York, 1990. (3) Hill, C. L. ActiVation and Functionalization of Alkanes; WileyInterscience: New York, 1989. (4) Sen, A. Acc. Chem. Res. 1988, 21, 421. (5) Shilov, A. E. ActiVation of Saturated Hydrocarbons by Transition Metal Complexes; D. Riedel Publishing Co.: Dordrecht, The Netherlands, 1984. (6) Waltz, K. M.; Hartwig, J. F. Science 1997, 277, 211. (7) Shilov, A. E.; Shul’pin, G. B. Chem. ReV. 1997, 97, 2879. (8) Goldberg, K. I.; Wick, D. D. J. Am. Chem. Soc. 1997, 119, 10235. (9) Bromberg, S. E.; Yang, H.; Asplund, C. M.; Lian, T.; McNamara, K. B.; Kotz, K. T.; Yeston, J. S.; Wilkens, M.; Frei, H.; Bergman, R. G.; Harris, C. B. Science 1997, 278, 260. (10) Hall, C.; Perutz, R. N. Chem. ReV. 1996, 96, 3125. (11) Periana, R. A.; Taube, D. J.; Gamble, S.; Taube, H.; Satoh, T.; Fuji, H. Science 1998, 280, 560.

10.1021/ja993015+ CCC: $19.00 © 2000 American Chemical Society Published on Web 02/18/2000

2042 J. Am. Chem. Soc., Vol. 122, No. 9, 2000 feasibility of mercuric salts as catalysts.12 Inspired by the successful experiments of Periana et al., we recently undertook a quantum chemical study of the thermochemistry of the cisplatin-catalyzed process in H2SO4, in an effort to test the thermodynamic feasibility of several possible reaction pathways using theoretical techniques.13 We found that the process proposed by Periana et al.,11 and also by Labinger, Bercaw, and their co-workers,14,15 where the crucial activation step occurs as an electrophilic attack on CH4 by a solvated [Pt(NH3)2(OSO3H)]+ molecule (T-complex), is thermodynamically feasible. The activation reaction is followed by oxidation of the resulting methyl complex to a bisulfate ester. While the homogeneous catalytic process of Periana et al.11 represents an obvious breakthrough, it would be obviously desirable if such a catalysis could be carried out in solvents that are more benign than concentrated H2SO4. Very recently Johansson et al. observed C-H activation under mild conditions in 2,2,2trifluoroethanol (TFE) at a cationic Pt(II) diimine aqua complex,16 which represents an obvious advance in this area. Our previous study was restricted to a study of the thermodynamics of the key reactions of the methane f methyl bisulfate ester conversion in sulfuric acid, and as such, it did not include the elucidation of full details of the reaction pathways, such as transition states and possible further intermediates. Given the large sizes of the mono- and bis(bisulfate) intermediates that are present in the reaction schemes considered, a detailed exploration of the potential energy surface represents a considerably more complex and resource-intensive study but one which we hope to complete soon. In an effort to gain useful insight and understanding of the factors that affect every step of the C-H activation process, we thought it instructive to carry out in the first instance a series of detailed studies on a simpler system: [PtCl2(NH3)2]. Siegbahn and Crabtree17 have recently made a theoretical study of the Shilov reaction involving [PtCl2(H2O)2], where in the initial reaction step a water ligand is found to be replaced by a CH4 molecule in the first coordination shell. In our studies of the analogous reaction schemes for the related [PtCl2(NH3)2] (cis and trans) an ammonia ligand is initially replaced by CH4. As the metal-ammonia bond in these complexes is expected to be considerably stronger than the metal-water bond in the aqua complexes, the activation energy for the transfer of CH4 into the first coordination shell is likely to be proportionately higher for the cis- and trans-platin reactionssin fact, this ligand exchange may not be the most energetically favorable reaction pathway. (As shown by the experiments of Periana et al.,11 as well as our previous theoretical work,13 cis-platin readily activates methane in sulfuric acid, where the first step is loss of chloride.) However, our interest is generally in an analysis of the effects of the nature of ligation on the reaction pathways and energetics, and in this work we are concerned with the theoretical comparison of H2O and NH3 replacement. A fundamental question is whether the reaction of CH4 with a metal complex leading to C-H bond breaking and formation of a metal-CH3 bond proceeds by electrophilic substitution (σ(12) Periana, R. A.; Taube, D. J.; Evitt, E. R.; Lo¨ffer, D. G.; Wentrecek, P. R.; Voss, G.; Masuda, T. Science 1993, 259, 340. (13) Mylvaganam, K.; Bacskay, G. B.; Hush, N. S. J. Am. Chem. Soc. 1999, 121, 4633. (14) Luinstra, G. A.; Wang, L.; Stahl, S. S.; Labinger J. A.; Bercaw, J. E. J. Organomet. Chem. 1995, 504, 75 (15) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. Angew. Chem., Int. Ed. Engl. 1998, 37, 2180. (16) Johansson, L.; Ryan, O. B.; Tilset, M. J. Am. Chem. Soc. 1999, 121, 1974. (17) Siegbahn, P. E. M.; Crabtree, R. H. J. Am. Chem. Soc. 1996, 118, 4442.

MylVaganam et al. bond metathesis) or oxidative addition, involving Pt(II) f Pt(IV) oxidation. For reaction of the trans-aqua complex in aqueous acid solution Siegbahn and Crabtree concluded that the most favored pathway was σ-bond metathesis, in which a proton is transferred from a methane σ-complex to a neighboring chloride ligand.17 This is in effect the mechanism suggested by Shilov.5,7 However, an oxidative addition pathway was found to be closely competitive and no firm distinction between these could be made on theoretical grounds. We find more complex behavior for the hypothetical cis- and trans-platin activation of methane by NH3 replacement, with some marked differences from those noted for the aqua systems. Interestingly, in their recent theoretical study of inter- and intramolecular C-H activation by a cationic iridium(III) complex, Niu and Hall found that the activation process occurs only by oxidative addition, via the formation of a Ir(IV) hydrido complex.18 In the course of this mechanistic study we also found that the products of an oxidative addition of CH4 to a Pt(II) complex would yield Pt(IV) methyl hydrido complexes with the methyl and hydride ligands effectively occupying cis positions relative to each other. Moreover, the comparative energetic studies that we also carried out have shown that placing these ligands trans to each other yields an energy that is ∼35 kcal mol-1 higher than for the corresponding cis isomer. Qualitatively, such an ordering in energy may be predicted on the basis of the strongly trans-directing natures of the hydride and methyl ligands,19,20 but the computed energy difference actually turned out to be larger than expected. Consequently, in light of this observation we have also recalculated the thermodynamics of the oxidative addition of methane to several potential Pt(II) catalysts considered in our previous work, placing CH3- and H- ligands cis to each other rather than trans as had been done before. Computational Methods As in our previous work, the calculations reported in this paper were performed using density functional theory (DFT), utilizing the hybrid functional B3LYP.21-24 The bulk of the calculations, in particular all geometry optimizations and vibrational frequency calculations, were carried out using the effective core potentials (ecp) and double-ζ basis sets of Stoll et al.25,26 for the heavy atoms and a double-ζ basis set for the H atoms.27 The resulting (valence) basis is thus a [6s5p3d] set on Pt, accommodating the 16 valence electrons, [2s2p] on C and N, and [2s3p] on Cl and [2s] on H, denoted ecp-DZ. The energies were then recalculated with a larger basis set, which is a full double-ζ plus polarization set28,29 on the C, N, Cl and H atoms (with all their electrons) but containing the same Pt ecp and basis as the smaller set described above. This larger basis is denoted ecp-DZP. The partition functions and hence thermal and zero point corrections to the molecular energies as well as molecular entropies were obtained using the standard formulas for quantum harmonic oscillators and classical rotors as well as classical translation, using the vibrational frequencies and rotational constants computed with the ecp-DZ basis. The solvation free energies were calculated using the isodensity polarized continum model (IPCM),30 which is a dielectric continuum (18) Niu, S.; Hall, M. B. J. Am. Chem. Soc. 1998, 120, 6169. (19) Hartley, F. R. Chem. Soc. ReV. 1973, 2, 163. (20) Lin, Z.; Hall, M. B. Inorg. Chem. 1991, 30, 646. (21) Becke, A. D. Phys. ReV. A 1988, 38, 3098. (22) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (23) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200. (24) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (25) Igel-Mann, G.; Stoll, H.; Preuss, H. Mol. Phys. 1988, 65, 1321. (26) Andrae, D.; Ha¨ussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Theor. Chem. Acta 1990, 77, 123. (27) Huzinaga, S. J. Chem. Phys. 1965, 42, 1293. (28) Dunning, T. H.; Hay, P. J. In Modern Theoretical Chemistry; Schaefer, H. F., III, Ed.; Plenum: New York, 1976; Vol. 3, p 1. (29) Frisch, M. J.; Pople, J. A.; Binkley, J. S. J. Chem. Phys. 1984, 80, 3265.

Homogeneous ConVersion of Methane to Methanol

J. Am. Chem. Soc., Vol. 122, No. 9, 2000 2043

type method, where the shape and size of the cavity occupied by the solute molecule is defined by an isodensity surface of the solute. Application of the IPCM method has, however, necessitated all electron calculations; these were performed using the ecp-DZ basis at the appropriate (gas phase) optimized geometries, the core orbitals of each atom expanded in terms of Huzinaga’s minimal basis sets.31 The charge distributions of the complexes were analyzed by the Roby-Davidson32-35 and the Mulliken methods.36 In the RobyDavidson approach the one-electron density is analyzed by defining and using a set of projection operators associated with each atom, as well as all pairs, triples, ..., of atoms, which enables the corresponding calculation of the number of electrons associated with individual atoms and pairs, triples, ..., of atoms. Generally, the most important quantities are the shared electron number between two atoms, which can be correlated with the degree of covalent bonding between those atoms and the net charges on the atoms. In the Mulliken method, the measure of electron sharing is provided by the overlap population. The methods and basis sets discussed above were used with a good degree of success in our previous work on platinum catalysts as well as molecular hydrogen complexes of osmium(II).37,38 The computations were carried out using the Gaussian94 and -98 programs39,40 using the DEC alpha workstations of the Theoretical Chemistry group at Sydney University.

Results and Discussion The reactions considered in this work are the replacement of an ammonia ligand by methane

PtCl2(NH3)2 + CH4 f PtCl2(CH4)(NH3)

(1)

followed by activation of the CH bond either by σ-bond metathesis yielding PtCl(HCl)(CH3)(NH3) or, oxidatively, by the formation of a Pt(IV) alkyl hydrido complex, viz. PtCl2(H)(CH3)(NH3). The final steps are reductive eliminations, where a proton is transferred to the ammonia ligand, so that NH4+ could be eliminated, and hence yielding the Pt(II) complex PtCl2(NH3)(CH3). In a study of an analogous set of reactions, namely those of the diaqua complex PtCl2(H2O)2, Siegbahn and (30) Foresman, J. B.; Keith, T. A.; Wiberg, K. B.; Snoonian, J.; Frisch, M. J. J. Phys. Chem. 1996, 100, 16069. (31) Huzinaga, S.; Andzelm, J.; Kobukowski, M.; Radzio-Andelm, E.; Sakai, Y.; Tatewaki, H. Gaussian Basis Sets for Molecular Calculations; Elsevier: Amsterdam, 1984. (32) Davidson, E. R. J. Chem. Phys. 1967, 46, 3320. (33) Roby, K. R. Mol. Phys. 1974, 27, 81. (34) Heinzmann, R.; Ahlrichs, R. Theor. Chim. Acta 1976, 42, 33. (35) Ehrhardt, C.; Ahlrichs, R. Theor. Chim. Acta 1985, 68, 231. (36) Mulliken, R. S. J. Chem. Phys. 1955, 23, 1833. (37) Bytheway, I.; Bacskay, G. B.; Hush, N. S. J. Phys. Chem. 1996, 100, 6023. (38) Bytheway, I.; Bacskay, G. B.; Hush, N. S. J. Phys. Chem. 1996, 100, 14899. (39) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.;. Zakrzewski, V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.;. Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; HeadGordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94 (ReVision E.2); Gaussian, Inc.: Pittsburgh, PA, 1995. (40) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. A.; Daniels, A. D.; Kudin, K. A.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Reploge, E. S.; Pople, J. A.; Gaussian 98 (ReVision A.3); Gaussian, Inc.: Pittsburgh, PA, 1998.

Crabtree17 found that, after methane exchanged with water, the presence of a solvating water molecule, bridging in effect the remaining water ligand and a chloride, has an important energetic effect. We have observed a similar effect in that the PtCl2(CH4)(NH3) complex could form a fairly strong bond with an ammonia or water molecule by way of a hydrogen bond to the metal-bound ammonia ligand. Therefore we decided to retain both ammonia molecules in our calculations, so the ammoniamethane exchange reaction is rewritten in the form

[PtCl2(NH3)2] + CH4 f [PtCl2(CH4)(NH3)]‚‚‚NH3 (2) The ligands inside the brackets are understood to be in the first coordination shell, with the second ammonia being effectively in the second coordination shell but hydrogen bonded to the ammonia ligand. We recognize of course that in an aqueous solution the second ammonia could be replaced by water, so we may well have [PtCl2(CH4)(NH3)]‚‚‚OH2. It is worth noting that ammonia, when bonded to a metal ion such as Pt2+, is quite acidic, i.e., it is a good proton donor, unlike free ammonia which is a very poor proton donor, as evidenced by the very weak H-bond in the ammonia dimer.41 Very strong hydrogen bonds of this type have been noted before, especially for systems with fluoride as a ligand. In a number of complexes it has been found that fluoride has the capacity to form intramolecular H- bonds with other ligands, such as NH2,42 or intermolecular H-bonds with other species, especially HF, in which case the system is best described as a complex of bifluoride.43,44 The computed (relative) electronic energies of the molecular species studied in this work that correspond to equilibrium and transition state geometries on the potential energy surfaces and all isomers of cis- and trans-[PtCl2(NH3)2]CH4 (as well as others studied in this work) along with their solvation energies, zero point energies, entropies, thermal corrections to the enthalpies, and the resulting relative enthalpies and free energies in aqueous solution at 298 K are given in Table 1. The total electronic energies (in Eh) are given in Table S1 of the Supporting Information. The structures that correspond to these species, viz. 1C-13C and 1T-16T, will be given in later sections of the paper that discuss the reaction pathways. The corresponding energetic information for the parent cis- and trans-platin complexes and free ligands is summarized in Table 2 and Table S1. Comparison of the relative ecp-DZ and ecp-DZP energies in Table 1 indicates that there is a reasonably good level of agreement between them, the differences being generally ∼3-5 kcal/mol, i.e., ∼10-20%, the ecp-DZP relative energies being consistently lower. The largest differences occur for the highest energy transition states and intermediate 6T-TS, 7T, and 8TTS. Since all the species in the reactions are neutral molecules, the solvation energies are fairly small (∼-10 kcal/mol), and thus, the solvent contributions to the reaction energies and barriers are quite modest. Exceptions are the transition state 6CTS and 8C-TS and the formation of the ion pair 13C. In the following sections we discuss the results obtained for the reactions between methane and cis- and trans-platin, respectively, as well as the relative stabilities of [PtCl2(NH3)2] and [PtCl2(H2O)2] and also the stabilities of the Pt(IV) isomers [PtCl2H(CH3)(NH3)2] and [Pt(NH3)2H(CH3)(OSO3H)2] and [Pt(NH3)2H(CH3)(OSO3H)(H2SO4)]. (41) Hassett, D. M.; Marsden, C. J.; Smith, B. J. Chem. Phys. Lett. 1991, 183, 449. (42) Patel, B. P.; Crabtree, R. H. J. Am. Chem. Soc. 1996, 118, 13105. (43) Murphy, V. J.; Hascall, T.; Chen, J. Y.; Parkin, G. J. Am. Chem. Soc. 1996, 118, 7428. (44) Whittlesey, M. K.; Perutz, R. N.; Greener, B.; Moore, M. H. Chem. Commun. 1997, 187.

2044 J. Am. Chem. Soc., Vol. 122, No. 9, 2000

MylVaganam et al.

Table 1. Platinum Complexes: Relative Gas-Phase Energies, Total and Relative Solvation Energies, Zero-Point Vibrational Energies, Thermal Contributions to Enthalpies, Standard Entropies, and Relative Enthalpies, and Gibbs Free Energies of the cis- and trans-Platin Based Intermediates and Products Obtained by the Ecp-DZ and Ecp-DZP Basis Sets (in Parentheses) at 298 Ka,b molecule

∆Eel(g)/ kcal mol-1

Esolv/ kcal mol-1

EZPE/ kcal mol-1

Htherm/ kcal mol-1

1C 2C-TS 3C 4C-TS 5C 6C-TS 7C 8C-TS 9C 10C-TS 11C 12C-TS 13C

0 37.0 (34.0) 35.6 (30.8) 35.9 (31.2) 26.8 (24.6) 49.8 (44.8) 36.4 (30.8) 35.1 (32.5) 29.3 (29.0) 31.9 (31.9) 14.8 (17.2) 38.7 (36.9) 7.5 (8.5)

-7.6 -7.1 -7.9 -10.8 -11.6 -18.1 -6.9 -15.0 -12.2 -11.6 -10.8 -9.5 -27.1

79.3 77.2 77.4 77.3 77.9 74.7 75.7 76.7 77.7 77.7 79.3 76.7 79.8

cis-Platin 88.4 86.2 86.4 86.0 86.6 82.9 84.1 84.7 86.0 85.5 87.0 85.1 87.9

1T 2T-TS 3T 4T-TS 5T 6T-TS 7T 8T-TS 9T-TS 10T 11T-TS 12T 13T-TS 14T 15T-TS 16T

0 45.5 (42.8) 43.2 (37.9) 45.0 (39.8) 31.0 (28.6) 72.2 (61.5) 70.5 (58.7) 73.0 (60.6) 43.5 (40.3) 42.3 (39.2) 43.7 (40.4) 25.7 (25.9) 48.1 (44.7) 32.8 (32.2) 62.0 (58.5) 38.9 (36.6)

-10.3 -8.6 -9.9 -10.8 -10.0 -8.6 -7.6 -10.2 -10.5 -8.7 -8.7 -12.1 -9.1 -11.7 -9.0 -10.3

79.6 77.3 77.8 77.1 78.2 74.3 75.4 75.1 76.6 76.8 76.8 78.8 77.4 77.7 75.5 77.6

trans-Platin 88.9 86.3 86.9 86.1 86.9 82.8 84.3 83.8 84.8 85.3 84.9 87.0 85.7 86.2 83.9 86.1

∆Eel(soln)/ kcal mol-1

∆H°298(soln)/ kcal mol-1

∆G°298 (soln)/ kcal mol-1

125.2 118.6 118.6 115.7 115.9 112.3 115.4 110.5 115.6 107.8 105.3 117.3 111.3

0 37.5 (34.5) 35.3 (30.5) 32.7 (28.0) 22.8 (20.6) 39.3 (34.3) 37.1 (31.5) 27.7 (25.1) 24.7 (24.4) 27.9 (27.9) 11.6 (14.0) 36.8 (35.0) -12.0 (-11.0)

0 33.3 33.3 30.3 21.0 33.8 32.8 24.0 22.3 25.0 10.2 33.5 -12.5

0 37.2 (34.2) 35.3 (30.5) 33.1 (28.4) 23.8 (21.6) 37.6 (32.6) 35.8 (30.2) 28.4 (25.8) 25.1 (24.8) 30.1 (30.1) 16.1 (18.5) 35.9 (34.1) -8.4 (-7.4)

126.9 120.2 118.6 118.4 117.0 115.0 120.0 115.4 112.9 114.6 111.3 112.0 112.1 113.5 113.9 116.4

0 47.2 (44.5) 44.9 (38.3) 45.4 (39.3) 31.3 (28.9) 73.9 (63.2) 73.2 (61.4) 73.1 (60.7) 43.3 (40.1) 43.9 (40.8) 45.3 (42.0) 23.9 (24.1) 49.3 (45.9) 31.4 (30.8) 63.3 (59.8) 38.9 (36.6)

0 44.6 42.9 42.6 29.3 67.8 68.6 68.0 39.2 40.3 41.3 22.0 46.1 28.7 58.3 36.1

0 46.6 (43.9) 45.3 (38.7) 45.1 (39.0) 32.2 (29.8) 71.3 (60.6) 70.7 (58.9) 71.3 (58.9) 43.3 (40.1) 44.0 (40.9) 46.0 (42.7) 26.4 (26.6) 50.5 (47.1) 32.6 (32.1) 62.1 (58.6) 39.2 (36.9)

S°/ cal K-1 mol-1

a The systems 1C-13C and 1T-16T represent reactants, products, intermediates, and transition states in the methane activation pathways. All are isomers of cis- and trans-[PtCl2(NH3)2]CH4. b Total enthalpies and Gibbs free energies (including solvation) are obtained as H°298(soln) ) Eel + therm therm therm n el el Esolv + Htherm 298 (g); G° 298(sol ) ) E + Esolv + G298 (g). H298 (g) and G298 (g) include the zero-point energy (EZPE) and E include nuclear repulsion.

Table 2. cis- and trans-Platin and Ligands: Solvation Energies, Zero-Point Vibrational Energies, and Thermal Contributions to Enthalpies and Standard Entropies at 298 K ligands

Esolv/ kcal mol-1

EZPE/ kcal mol-1

Htherm/ kcal mol-1

S°/ cal K-1 mol-1

cis-platin trans-platin ClNH3 CH4 H2O

-9.4 -12.9 -63.2 -3.6 -0.1 -7.1

50.5 50.7

56.7 56.6

95.4 96.7

20.7 28.0 12.3

23.2 30.4 14.7

48.5 46.7 45.2

Relative Stabilities of cis- and trans-Platin. According to our computations, trans-platin is more stable than its cis isomer, in both gas phase and aqueous solution, by ∼14 and ∼18 kcal/ mol, respectively. This is in qualitative agreement with the available experimental data on the relative heats of formation of cis- and trans-platin in their respective standard (crystalline) states, according to which trans-platin is more stable by 3 kcal/ mol.45 There appears to be no such data on the relative stabilities of these isomers in aqueous solution. Given that trans-platin is nondipolar, in contrast with a computed dipole of 12.3 D for cis-platin, it may seem somewhat surprising that the trans isomer appears to be more highly solvated. The situation is analogous to that noted for the cis and trans isomers of [PdCl2(H2O)2], where Siegbahn and Crabtree46 found in their self-consistent reaction field (SCRF) (45) Chernyaev, I. I.; Palkin, V. A.; Baranova, R. A.; Kuz’mina, N. N. Russian J. Inorg. Chem. 1960, 5, 693.

calculations that the larger solvation energy of the trans isomer is due mostly to the large difference in the quadrupole moments, with nonnegligible octupole and hexadecapole contributions. In our work, use of the IPCM-polarized continuum method effectively bypasses the potential convergence problem of the multipole expansion method but it neglects the nonelectrostatic contributions to the solvation such as dispersion, repulsion, and cavitation.47 In a further study of the solvation of cis- and trans-platin we carried out calculations on these molecules in the presence of four discrete water molecules, optimizing the gas-phase geometries of the resulting [PtCl2(NH3)2](H2O)4 clusters at the B3LYP/ecp-DZ level of theory, which was followed by the computation of the solvation energy of each cluster by the IPCM method. The structures obtained are shown in Figure 1. We expected the water molecules to bridge the ammonia and chloride ligands via hydrogen bonds, but such a structure was only obtained for trans-platin (structure A). In the case of cisplatin the optimizations consistently yielded a structure with just one water bridging the two ammonias as well as forming a H-bonded chain with the remaining three water molecules, as shown also in Figure 1. Evidently, the water-water hydrogen bonds are considerably stronger than the alternative waterchloride bonds, etc. Indeed, an alternative trans-platin based isomer (structure B) was subsequently found, where a given (46) Siegbahn, P. E. M.; Crabtree, R. H. Mol. Phys. 1996, 89, 279. (47) Bacskay, G. B.; Reimers, J. R. In Encyclopedia of Computational Chemistry; Schleyer, P. v. R., Ed.; J. Wiley & Sons Ltd.: Chichester, U.K., 1998; p 2620.


J. Am. Chem. Soc., Vol. 122, No. 9, 2000 2045 Table 3. Gas-Phase Binding Energies (in kcal/mol) of Ligands in Various Pt(II) Complexes Obtained at the B3LYP/ecp-DZ Level of Theory (without Zero-Point Vibrational Corrections) ligands

b

complex

NH3

Cl-

cis-PtCl2(NH3)2 trans-PtCl2(NH3)2 cis-PtCl2(H2O)2 trans-PtCl2(H2O)2

51.5 64.4

151.0 154.4 166.6 158.6

H2O

CH4

32.1 46.4

7.3a 14.5a 8.6b 21.6b

a Binding energy of methane in cis- and trans-PtCl2NH3CH4. Binding energy of methane in cis- and trans-PtCl2H2OCH4.

between cis- and trans-platin is reduced to ∼5 kcal/mol, in favor of the cis isomer These results, although qualitatively different from the initial IPCM estimates of 3.5 kcal/mol in favor of transplatin (see Table 1), would not significantly alter our earlier conclusion that the trans isomer is expected to be more stable in solution. Clearly, however, a significantly larger number of discrete water molecules are needed to adequately model the solvation of these simple platinum complexes. While we cannot rule out the possibility that a more accurate treatment of solvation would yield qualitatively different results, we believe that it is unlikely that the change would be large enough to result in a substantial change in the total cis-trans energy difference, most of which is electronic in origin. Relative Stabilities of the Pt(II) Diammine and Diaqua Complexes. The Shilov reaction occurs for platinum salts in aqueous acid solution, where the active metal complexes are believed to be diaqua Pt complexes such as [PtCl2(H2O)2]. If the rate-determining step in the alkyl activation reactions is the exchange of methane with an inner-sphere ligand such as water or ammonia, the rate of the overall reaction could be, to a large extent, determined by the endothermicity of that exchange reaction, which will strongly depend on the binding energy of H2O or NH3 in the appropriate complex. These were obtained by computing the (gas phase) energies of the dissociation reactions

[PtCl2X2] f [PtCl2X] + X X ) H2O, NH3

Figure 1. Structures of cis- and trans-[PtCl2(NH3)2](H2O)4 clusters.

pair of ammonia and chloride ligands are bridged by a water dimer, and which has a lower energy. The gas-phase binding energies of the four solvating water molecules, defined as the energies of the reactions

[PtCl2(NH3)2] + 4H2O f [PtCl2(NH3)2](H2O)4

(3)

were computed (at the ecp/DZP level, without zero point corrections) as -54.9, -41.5, and -45.9 kcal/mol for the cis, trans (A), and trans (B) isomers, respectively. These results suggest that the cis isomer may be better solvated, by up to 9 kcal/mol, as a result of the stronger hydrogen bonds in the former. However, when the secondary solvation effects with the bulk solvent are also calculated, using the IPCM method, the reverse trend is obtained. The computed solvation energies of the three clusters are -10.5, -16.5, and -14.2 kcal/mol. When these results are combined with those from the gas-phase cluster calculations, the difference in overall solvation energies

(4)

The results are given in Table 3, along with the binding energies of CH4 (in [PtCl2(CH4)X]) and Cl-, which had also been calculated in an analogous fashion. Clearly, ammonia is more tightly bound than water, the difference in their binding energies being ∼19 kcal/mol. Therefore, the activation energies for the methane-ammonia exchange are expected to be roughly that much higher than for the analogous methane-water exchange. Interestingly, the binding energy of methane depends to a significant degree on the other ligands in the complex. Thus in trans-[PtCl2(CH4)(H2O)] methane is computed to be bound by 21.6 kcal/mol, to be compared with 7.3 kcal/mol in cis-[PtCl2(CH4)(NH3)]. Therefore the low barrier to methane-water exchange in trans-[PtCl2(H2O)2] is probably partly due to the correspondingly low endothermicity of that reaction. Alternatively, considering the energetics of the substitution reaction

[PtCl2(NH3)2] + 2H2O f [PtCl2(H2O)2] + 2NH3

(5)

we computed (gas phase) reaction energies of 43.7 and 52.7 kcal/mol for the cis and trans isomers, respectively. These results, in good agreement with the estimates discussed above, indicate that ammonia is bound by ∼24 kcal/mol more tightly than water.


MylVaganam et al.

Figure 2. Schematic potential energy surface for methane activation by cis-platin.

Figure 3. CH4/NH3 exchange in cis-platin: Structures of intermediates and transition states.

Methane Activation by cis-Platin. Two distinct reaction pathways were found in the study of the PES of this reaction, as shown by the diagram in Figure 2. The initial reactant molecule is cis-platin (1C) with a methane molecule in the second coordination shell. The first reaction is the exchange of this methane with ammonia, producing a complex (3C) where methane is now in the inner coordination shell as ligand, while the ammonia, having moved to the second coordination shell, is only weakly bound. The structures associated with this first step are shown in Figure 3. As may be expected, migration of this ammonia molecule so as to hydrogen bond to the ammonia ligand is a facile process, with a very small or no barrier at all. While a small barrier corresponding to 4C-TS was located at the DFT/ecp-DZ level of theory, as the results in the tables indicate, once solvation and entropic effects are included, the

energy barrier located on the gas-phase PES disappears altogether. The net reaction for this first phase is then

[PtCl2(NH3)2](CH4) f [PtCl2(CH4)(NH3)]‚‚‚NH3 1C 5C (via 2C-TS, 3C, 4C-TS) (6) where the highest free energy barrier, computed as 34.2 kcal/ mol, corresponds to the transition state 2C-TS for the initial methane-ammonia exchange. By contrast, the barrier for the exchange of methane with water for cis-[PtCl2(H2O)2] has been estimated as just 12.8 kcal/mol. CH activation now can occur in two distinct ways. In the σ-bond metathesis proton migration occurs to the chloride cis to methane (see structures in Figure 4) but in such a way that



the proton ends up on the far side of the chloride, so that it bridges the two chlorides, i.e., giving rise to a bichloride:

[PtCl2(CH4)(NH3)]‚‚‚NH3 f 5C [Pt(Cl‚‚‚H‚‚‚Cl)(CH3)(NH3)]‚‚‚NH3 (via 6C-TS) (7) 7C In contrast with the observed bifluoride complexes,42-44 the bichloride ion in 7C acts as a bidentate ligand. In the transition state the distance between the migrating hydrogen and the methyl carbon is ∼3.4 Å, suggesting that the CH bond must be completely broken. Moreover the hydrogen-platinum distance is only 1.85 Å. It was confirmed however that the normal coordinate corresponding to the imaginary frequency (454i cm-1) effectively describes a large amplitude oscillation of the hydrogen between the carbon and chlorine atoms. Once the product 7C is formed, it could be deprotonated by a solvent molecule, or by the H-bonded ammonia, resulting in the formation of [PtCl2H(CH3)(NH3)]- and NH4+ or the ion pair 13C. An alternative, possibly more obvious pathway, where the migrating hydrogen would end up between the methyl and the chloride cis to it, so that it would be bonded to the latter, was also considered. However, as the resulting product was found to lie ∼13 kcal/mol above 7C, we did not attempt to locate the transition state associated with this process. The alternative pathway to σ-bond metathesis is the oxidative addition of methane to cis-platin (see structures 8C-TS-13C in Figure 5). In the first step hydrogen migrates to the platinum ion, effectively oxidizing it to Pt(IV):

[PtCl2(CH4)(NH3)]‚‚‚NH3 f [PtCl2H(CH3)(NH3)]‚‚‚NH3 5C 9C (via 8C-TS) (8) The intermediate 9C is now a pentacoordinate Pt(IV) complex, best described as a trigonal bipyramid, with the ammonia dimer and a chloride occupying the axial positions. It could become hexacoordinate by a solvating water molecule becoming a ligand or by utilizing the nearby ammonia in the second coordination shell. We considered both possibilities. The latter corresponds to the reaction

[PtCl2H(CH3)(NH3)]‚‚‚NH3 f [PtCl2H(CH3)(NH3)2] 9C 11C (via 10C-TS) (9) which results in an (approximately) octahedral Pt(IV) complex (11C) with the methyl and ammonia ligands occupying the axial positions with the hydride being cis to the equatorial ammonia. The alternative, where a water molecule exchanges with an ammonia and moves into the first coordination shell, i.e.

[PtCl2H(CH3)(NH3)]‚‚‚NH3 + H2O f [PtCl2H(CH3)(NH3)(H2O)] + NH3 (10) was found to be 9.2 kcal/mol less favorable than reaction 8. Given 11C, the final step is a highly concerted reaction that enables Pt(IV) to revert to Pt(II), while the platinum-axial ammonia bond is broken as the hydrogen, as proton, migrates to the leaving ammonia

[PtCl2H(CH3)(NH3)2] f [PtCl2(CH3)(NH3)]-(NH4+) 11C 13C (via 12C-TS) (11)

Figure 4. C-H activation by σ-bond metathesis in cis-platin: Structures of intermediates and transition states.

resulting in a highly stable ion pair (13C), where an ammonium ion is hydrogen bonded to the two chloride ligands. A hypothetical alternative to this process is the concerted, onestep isomerization of 9C to 13C, corresponding to a simultaneous migration of the hydrogen and ammonia to the chlorines. However, the barrier for this reaction was found to be ∼33 kcal/ mol (relative to 9C), which effectively rules out this type of reaction. The dissociation energy of the ion pair was computed to be 25.1 kcal/mol in solution. Given the huge solvation contribution (∼100 kcal/mol) to this energy, and a conservative estimate of its accuracy as ∼10%, the above binding energy could well be an overestimate. Thus the ionization of 13C is considered a definite possibility. In summary, the barriers associated with the methaneammonia exchange and the subsequent C-H activation and proton elimination steps are comparable, as are the barriers for the σ-bond metathesis and oxidative addition pathways. Thus once methane is in the first coordination shell, the reaction is predicted to be effectively free energy neutral. Methane Activation by trans-Platin. The PES obtained for the methane activation by trans-platin is shown in Figure 6. In comparison with cis-platin, the initial methane-ammonia exchange is a considerably more energetic process, with the trans-platin transition state 2T-TS corresponding to a free energy barrier of ∼44 kcal/mol, viz. ∼10 kcal/mol more than for cisplatin. In analogy with cis-platin, the product of this first reaction is [PtCl2(CH4)(NH3)]‚‚‚NH3 (5T), which is ∼8 kcal/mol less stable than the cis isomer 5C, relative to the respective parent species [PtCl2(NH3)2]CH4 (1T or 1C). The structures corresponding to this first step are shown in Figure 7. The σ-bond metathesis pathway (see structures in Figure 8)


MylVaganam et al.

Figure 5. C-H activation by oxidative addition in cis-platin: Structures of intermediates and transition states.

Figure 6. Schematic potential energy surface for methane activation by trans-platin.

for the activation of a C-H bond of methane appears to be much less favorable than in the case of cis-platin. As a proton migrates to one of the chloride ligands, a barrier with a free energy of ∼31 kcal/mol (relative to 5T) is encountered, corresponding to a 4-center transition state (6T-TS). The free energy of the resulting intermediate, viz. [Pt(HCl)Cl(CH3)(NH3)]‚‚‚NH3 (7T), is only marginally lower than that of the transition state 6T-TS. In contrast with these results, the analogous transition state and product for hydrogen migration in the trans-[PtCl2(H2O)2] system were found to be just 16.5

and 6.5 kcal/mol, respectively, above the energy of trans-[PtCl2(H2O)(CH4)](H2O) in the study of Siegbahn and Crabtree.17 Furthermore, they believe that the next step, where the proton is transferred from the chloride to the water ligand, is a facile process that results in the formation of [PtCl2(CH3)]-(H5O2)+, where the H5O2+ moiety is hydrogen bonded to the chlorides. We were unable to find a transition state that would lead to the analogous system, viz. [PtCl2(CH3)]-(N2H7)+. The optimizations consistently yielded a transition state (8T-TS) that leads to a 5-coordinate square pyramidal Pt(IV) hydride, [PtCl2H(CH3)-



Figure 7. CH4/NH3 exchange in trans-platin: Structures of intermediates and transition states.

(NH3)]‚‚‚NH3 (10T), which is only ∼11 kcal/mol higher in energy than [PtCl2(CH4)(NH3)]‚‚‚NH3 (5T). Interestingly, the analogous intermediate to 10T in the cis-platin activation pathway, viz. 9C, has a trigonal bipyramid structure. Although 7T, with a Pt-ClH bond, may appear somewhat unusual, we note that an iridium complex with an Ir-FH bond has been experimentally observed by Patel and Crabtree.42 On exploring the oxidative addition mechanism (see structures in Figure 9), we found that there is a low-energy barrier (9TTS) that directly leads to [PtCl2H(CH3)(NH3)]‚‚‚NH3 (10T). The intermediate 10T can be further stabilized by allowing it to become hexacoordinate, e.g. by moving the ammonia in the second coordination shell into the appropriate axial position. The resulting trans-[PtCl2H(CH3)(NH3)2] molecule (12T) is in fact the most stable system on the PES. Thus, for trans-platin both the σ-bond metathesis and oxidative addition mechanisms will ultimately yield an oxidized Pt(IV) intermediate. The reduction of 12T to [PtCl2(NH3)(CH3)]-(NH4)+ (14T) (see structures in Figure 10) is an endothermic process that can take place by returning to the 5-coordinate intermediate 10T, followed by proton migration via the transition state 13T-TS. Alternatively, 10T can rearrange to yield the complex 16T that can be regarded as a [PtCl2(CH3)]-(N2H7)+ ion pair, although this is a considerably more energetic process. 16T is analogous to the product [PtCl2(CH3)]-(H5O2)+ identified by Siegbahn and Crabtree in their work on the trans diaqua complex.17 In summary, as in the cis-platin system, the methaneammonia exchange and C-H activation and proton-transfer steps have comparable barriers. In an absolute sense however the barriers are ∼12 kcal/mol higher than for cis-platin. Moreover, in the trans-platin system the C-H activation process is predicted to take place predominantly by oxidative addition, with the σ-bond metathesis being no longer competitive with the oxidative mechanism. The catalysis of methane activation by platins is therefore qualitatively different from that which Siegbahn and Crabtree17

found for the Shilov reaction where the catalyst is [PtCl2(H2O)2] and where the reactions of the cis and trans isomers have comparable energetics, with C-H activation, either by σ-bond metathesis or oxidative addition, being the most energetic steps. Charge Distribution and Bonding. In an effort to better understand the chemistry of C-H activation in terms of simple bonding ideas, we carried out Roby-Davidson and Mulliken population analyses for the parent compounds, cis- and transplatin, as well as selected intermediates and products that contribute to the reaction pathways. The results are summarized in Table 4. Although the two methods yield qualitatively similar results, especially with respect to the trends, there are obvious major differences. The Roby-Davidson method assigns a positive charge of ∼0.9-1.3 e to Pt, while the corresponding Mulliken charges range from ∼-0.30 to -0.10. Given that formally, at least, platinum in these molecules has an oxidation state of II or IV, an overall negative charge is most likely an artifact. Therefore, as in our previous work on osmium complexes,48 we prefer to use the Roby-Davidson method to aid our understanding of bonding patterns in terms of simple, chemical concepts. Methane, as ligand, appears essentially neutral, while methyl is negative, with a charge that varies between -0.13 and -0.37 e. Interestingly, methyl is most negative in the least stable trans systems 7T and 10T. The ammonia moieties appear effectively neutral, but the charge on the chlorines is mostly ∼-0.60 e, except where they are bonded to the hydrogen, as in 7C and 7T. As one may expect, the chlorides represent the largest source of charge donation to Pt. The shared electron numbers between two atoms are interpreted as an indication of the covalency of the bond between the atoms. The largest degree of electron sharing occurs between Pt and C in the methyl complexes and between Pt and H in the hydrides, and therefore, only those results are quoted in Table 4. On the basis of this and the computed charges on the methyl and (48) Craw, J. S.; Bacskay, G. B.; Hush, N. S. J. Am. Chem. Soc. 1994, 116, 5937.


MylVaganam et al.

Figure 8. C-H activation by σ-bond metathesis in trans-platin: Structures of intermediates and transition states.

hydride ligands, we may conclude that the Pt-CH3 and Pt-H bonds are largely covalent. By contrast, there is considerably less electron sharing between the Pt and the chloride or ammonia ligands. Thus, there appear to be two distinct types of bonding mechanisms: covalent bonding, as characterized by the PtCH3 and Pt-H bonds, and largely electrostatic crystal field type interactions with little covalent character, as in the case of the “classical” ligands NH3 and Cl-. The pattern in the shared electron numbers is roughly reflected in the Mulliken overlap charges. In summary, while describing the methyl hydrido complexes in terms of Pt4+ and CH3- and H- ligands is formally correct, the chemically more accurate description would be in terms of covalent bonds between Pt2+ and CH3 and H radicals. On the basis of the above analysis, the Pt(IV) methyl hydrido complexes could be described in terms of a covalently bonded (H-Pt-CH3)2+ molecule which is able to coordinate up to four ligands, such as chloride and ammonia. As in analogous Pd complexes,46 such covalent bonding would utilize sd type hybrids of Pt2+, which would result in near 90° bond angles. Recalling that the hydride and methyl anions as ligands are known to have strongly trans-influencing and trans-influenced

Figure 9. C-H activation by oxidative addition in trans-platin: Structures of intermediates and transition states.

properties, i.e., a strong preference for not being trans to each other, such behavior may well be a consequence of the covalent bonding between Pt2+ and the H and CH3 radicals. Relative Stabilities of Isomers of Pt(IV) Methyl Hydrido Complexes. The structures of the Pt(IV) methyl hydrido complexes identified as intermediates in this work (9C, 11C, 10T, and 12T) are such that the methyl and hydride ligands are cis to each other, inasmuch as one is equatorial while the other is axial. However, in our previous work, when exploring the reaction energies associated with oxidative addition mechanisms, where methane adds on to Pt(NH3)2(OSO3H)2, Pt(NH3)2(OSO3H)(H2SO4), or indeed Pt(NH3)2(Cl)2, the methyl and


J. Am. Chem. Soc., Vol. 122, No. 9, 2000 2051 the strongly trans-directing natures of these two ligands, the resulting trans Pt(IV) complexes would be expected to be significantly less stable than the cis isomers. To quantify the importance of such trans influence in the context of methane activation, a detailed comparison of the relative stabilities of such cis and trans isomers has also been carried out, for both the current systems, as well as those studied in our previous work. The results are summarized in Table 5. The energies of the four isomers where CH3 and H are cis to each other are within ∼7 kcal/mol of each other, but they are ∼32-40 kcal/mol lower than the energy of the fifth isomer where CH3 and H are trans to each other, as expected on the basis of trans influence. The variations in energy among the bisulfate complexes are similar to that for the simple chlorides, except where the bisulfates are trans to each other, with CH3 and H cis. The higher energy of this particular isomer can be attributed to the absence hydrogen bonding between bisulfates in this trans complex. The relative stabilization that results from hydrogen bonding in molecules where the bisulfates are cis to each other was estimated by comparing the energies of the bisulfate dimer moieties at the geometries they assume in the first two isomers from Table 5. The energy of the trans configuration was computed to be 22.9 kcal/mol higher than for the cis. This is in near-perfect agreement with the energy difference of 20.1 kcal/mol that had been obtained for the corresponding Pt complexes. In light of this study we conclude that contrary to our earlier findings, the oxidative addition type reactions

[Pt(NH3)2(OSO3H)2] + CH4 f [Pt(CH3)H(NH3)2(OSO3H)2] (12) and

[Pt(NH3)2(OSO3H)(H2SO4)]+ + CH4 f [Pt(CH3)H(NH3)2(OSO3H)(H2SO4)]+ (13) could be thermodynamically viable. Using the lowest energy structures from Table 5 for the above methyl hydrido complexes (with the ammonias trans to each other), the reaction enthalpies in solution at 298 K for reactions 12 and 13 were computed to be 5.2 and 10.4 kcal mol-1, respectively. The corresponding reaction free energies are 17.1 and 20.6 kcal mol-1, while at 473 K and a pressure of 500 psi the free energies are 20.7 and 23.2 kcal mol-1. Conclusion

Figure 10. Reduction of trans-[PtCl2H(CH3)(NH3)2]: Structures of intermediates and transition states.

hydride ligands were placed trans to each other as axial ligands. As pointed out by Labinger,49 and also discussed above, given

In our study of C-H activation of methane catalyzed by cisand trans-platin in aqueous solution, where the initial step is the replacement of an ammonia ligand by methane, we found that once that substitution has taken place, the barriers to the subsequent activation and proton elimination steps are comparable with those of the initial substitution reaction. However, in contrast with the Shilov reaction studied by Siegbahn and Crabtree,17 the initial energy barriers are substantially higher in the case of the platins, as are the overall heats and free energy changes of the overall substitution reactions. This can be traced to the relatively strong platinum-ammonia bond. In the case of cis-platin the C-H activation is expected to occur by both σ-bond metathesis and oxidative addition. The energy barriers associated with these two distinct pathways were found to be comparable. However, in the trans-platin system the C-H activation process is predicted to take place predominantly by (49) Labinger, J. A. Private communication.


MylVaganam et al.

Table 4. Results of Roby-Davidson (RD) and Mulliken Population Analyses on Selected Reactants, Intermediates, and Products: RD and Mulliken Charges on Pt and Ligands and RD Shared Electron Numbers and Mulliken Overlap Populations (Mulliken Results in Parentheses) RD shared electron nos. (Mulliken overlap populations)

RD and (Mulliken) charges on Pt and ligands Pt cis-platin 5C 7C 9C (IV) 13C trans-platin 1T 5T 7T 10T (IV) 14T

0.96 (0.05) 0.95 (-0.24) 0.85 (-0.23) 1.28 (-0.13) 0.85 (-0.15) 0.98 (-0.09) 0.95 (-0.14) 0.97 (-0.31) 0.85 (-0.31) 1.29 (-0.11) 0.99 (-0.58)

CH3 or CH4 0.05 (0.40) -0.17 (0.15) -0.13 (0.30) -0.20 (0.10) 0.08 (0.04) 0.08 (0.47) -0.37 (0.10) -0.25 (0.30) -0.37 (0.22)

NH3 or NH4+

Cl

0.11 (0.40) 0.10 (0.17), 0.03 (0.42) 0.14 (0.16), 0.05 (0.40) 0.15 (0.17), 0.01 (0.44) 0.63 (0.71), 0.15 (0.42) 0.11 (0.49) 0.03 (0.49), 0.14 (0.49) 0.11 (0.17), 0.07 (0.51) 0.09 (0.14), 0.03 (0.36) 0.09 (0.15), 0.00 (0.37) 0.74 (-0.47), 0.10 (0.38)

-0.60 (-0.42) -0.60 (-0.38), -0.53 (-0.37) -0.55 (-0.35), -0.41 (-0.12) -0.60 (-0.32), -0.72 (-0.47) -0.68 (-0.47), -0.77 (-0.62) -0.62 (-0.44) -0.59 (-0.43), -0.62 (-0.45) -0.64 (-0.43), -0.60 (-0.41) -0.65 (-0.44), -0.13 (+0.15) -0.65 (-0.37), -0.62 (-0.34) -0.74 (-0.47), -0.71 (-0.43)

H

Pt/C

0.03 0.00

0.06 (0.16) 0.54 (0.35) 0.43 (0.34) 0.52 (0.36)

0.07 0.01 (-0.05) 0.65 (0.27) 0.00 (-0.01)

0.00 (0.01) 0.07 (0.20) 0.46 (0.39) 0.39 (0.36) 0.43 (0.39)

0.08 (0.08) -0.01 (-0.03) 0.69 (0.25) 0.00 (-0.01)

0.19 0.14

Pt/H

Table 5. Relative Gas-Phase Energies (without Zero-Point Vibrational Corrections) of Isomers of [Pt(X)(NH3)2(H)(CH3)] (X ) Cl, OSO3H) Computed at the B3LYP/ecp-DZ Level of Theory (in kcal/mol)

and methyl anions, and hence their preference to be cis to each other, may be associated with the tendency of the parent radicals to form covalent bonds with Pt2+. In light of the above findings the energies of several Pt(IV) methyl hydrido bisulfate complexes were also recalculated. The new results provide evidence for the thermodynamic feasibility of oxidative addition of methane to catalysts such as [Pt(NH3)2(OSO3H)2] or [Pt(NH3)2(OSO3H)(H2SO4)]+.

oxidative addition, as the σ-bond metathesis was found to be associated with a much higher barrier, in further contrast with the Shilov reaction where the two pathways were found to be comparable with regard to their activation energies. A valuable finding of our study is that, in line with accepted ideas on trans influence, the methyl and hydride ligands in the Pt(IV) complexes that arise in the oxidative addition reactions are always cis to each other. However, the results of the Roby-Davidson analyses suggest that the Pt-H and Pt-CH3 bonds are best described as covalent bonds. The possible consequence of this is that the strongly trans-influencing and trans-influenced properties of the hydride

Acknowledgment. We are pleased to acknowledge the helpful comments of Professor Henry Taube (Stanford University) and also those of Professor J. A. Labinger (California Institute of Technology) concerning the trans-directing properties of methyl and hydride ligands and their importance in oxidative addition mechanisms. Supporting Information Available: Table S1, containing the total electronic energies (in Eh) of all molecular species studied in this work (PDF). This material is available free of charge via the Internet at http://pubs.acs.org. JA993015+

Homogeneous Conversion of Methane to Methanol. 2. Catalytic

Recommend Documents